Prof. IL Jeon of SKKU Advanced Institute of NanoTechology (SAINT) at Sungkyunkwan University (SKKU), alongside Dr. Sihyeok Kim from Prof. Jeon’s group, has successfully developed a high-sensitivity acetylene (C2H2) gas sensor by embedding carbon nanotubes (CNTs) within a polyimide (PI) matrix. This sensor demonstrates stable operation without delamination, even under prolonged exposure to oil environments.

Conventional metal-oxide-based gas sensors, commonly used for hydrocarbon gas detection, exhibit significant limitations in oil-based environments due to the extremely low oxygen content (less than 2%) in oil compared to atmospheric conditions. Furthermore, these sensors require high operating temperatures exceeding 300 °C and often struggle to distinguish between hydrocarbon gases with similar molecular structures.

CNT-based sensors overcome these limitations through their large surface area and weak π-π interactions, enabling gas detection via physical adsorption driven by van der Waals forces, rather than relying solely on oxidation-reduction reactions. As a result, these sensors can operate effectively in oxygen-deficient oil environments while maintaining superior detection performance at the average transformer oil operating temperature of 90 °C.

However, conventional CNT-based sensors face practical challenges, such as CNT delamination under prolonged environmental fluctuations, including mechanical vibrations and oil convection. Additionally, residual surfactants and nanotube bundling reduce the effective surface area of the sensing layer, impairing sensor sensitivity, response speed, and recovery time.

Prof. Jeon’s research team has the FCCVD (Floating Catalyst Chemical Vapour Deposition)-based CNT thin-film technology exclusively in Korea, recognised globally for its excellence. With over a decade of continuous research in this area, the team successfully embedded high-quality CNT films into PI thin films, achieving sensors that maintain structural integrity and operational stability even under extended exposure to oil environments. Six months prior, the team published their world-class flexible CNT sensor based on PI in the prestigious journal Advanced Materials (IF: 29.4, https://doi.org/10.1002/adma.202313830).

Figure 1. (a) Fabrication process of the Au-CNT C2H2 gas sensor embedded in PI (b) Image of the fabricated sensor

To further enhance performance, the team integrated a multi-layer heater beneath the sensor layer to maintain the optimal surface temperature of 90 °C. The upper sensor layer was constructed using a thin CNT film with high gas permeability to maximise sensitivity, while the lower heater layer utilised a thicker CNT film to optimise heat distribution. The sensor demonstrated an exceptional response of approximately 10.4% when exposed to a 30 ppm acetylene concentration. Moreover, it exhibited rapid response and recovery times of 444 and 670 seconds, respectively, even under oil immersion conditions.

Figure 2. (a) Sensor response characteristics as a function of temperature (b) Response and recovery times at varying temperatures (c) Sensitivity and linearity across 5-100 ppm concentrations (d) Repeatability at 90 °C (e) Hysteresis performance

The research team confirmed that embedding FCCVD-derived CNTs into PI thin films effectively prevents delamination and ensures long-term operational stability in oil environments. They highlighted the potential for further performance enhancements through optimised CNT synthesis and integration with AI-driven analytical frameworks.

This groundbreaking research was published on November 28 in Advanced Materials (IF: 29.4; https://doi.org/10.1002/adma.202410179), marking a significant achievement with two consecutive publications within six months.

The project received support from the Ministry of Science and ICT, the National Research Foundation of Korea, and Prof. Jeon’s startup, JLabNT co ltd.

※ Paper Title: Highly Sensitive and Stable In Situ Acetylene Detection in Transformer Oil Using Polyimide-Embedded Carbon Nanotubes

※ Original Paper:

https://advanced.onlinelibrary.wiley.com/doi/full/10.1002/adma.202410179